# DeepSeek-V3.1-Terminus vs Magistral Medium

> DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 34.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-terminus-vs-magistral-medium
- Last updated: 2026-10-10
- Shared benchmarks: 13

## Summary

- They share 13 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Magistral Medium in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V3.1-Terminus leads 25.7 to 8.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 57.4% for DeepSeek-V3.1-Terminus and 16.2% for Magistral Medium.
- DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 per million input/output tokens, against $2 / $5 for Magistral Medium.
- DeepSeek-V3.1-Terminus accepts more context: 164K tokens versus 128K.

## Snapshot

| | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 42.6 | 34.8 |
| Rank | 96 | 224 |
| Context | 164K | 128K |
| Input $/M | $0.30 | $2 |
| Output $/M | $1 | $5 |
| Weights | Open | Open |

## Coding

- DeepSeek-V3.1-Terminus: 41.5 (#114)
- Magistral Medium: 38.6 (#161)

| Benchmark | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| SciCode | 40.6% | 39.2% |
| LMArena Coding | 1426 | 1319 |
| ALE-Bench | 745.17 | — |

## Reasoning

- DeepSeek-V3.1-Terminus: 25.7 (#126)
- Magistral Medium: 8.7 (#342)

| Benchmark | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 57.4% | 16.2% |
| CritPt | 1.7% | 0.3% |
| LMArena Hard Prompts | 1427 | 1267 |
| ARC-AGI-2 | — | 0% |
| ARC-AGI-1 | — | 6.1% |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |

## Math

- DeepSeek-V3.1-Terminus: 35.7 (#138)
- Magistral Medium: 32.4 (#187)

| Benchmark | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| LMArena Math | 1402 | 1249 |

## Knowledge

- DeepSeek-V3.1-Terminus: —
- Magistral Medium: 32.9 (#199)

| Benchmark | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| LMArena Expert | — | 1222 |

## Multilingual

- DeepSeek-V3.1-Terminus: 52.3 (#87)
- Magistral Medium: 39.8 (#219)

| Benchmark | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| LMArena Non-English | 1407 | 1232 |
| LMArena Russian | 1434 | 1224 |
| LMArena Chinese | — | 1225 |
| LMArena French | — | 1268 |
| LMArena German | — | 1248 |
| LMArena Japanese | — | 1176 |
| LMArena Korean | — | 1125 |
| LMArena Spanish | — | 1269 |

## Instruction Following

- DeepSeek-V3.1-Terminus: 74.4 (#96)
- Magistral Medium: 66.2 (#207)

| Benchmark | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| LMArena Instruction Following | 1406 | 1254 |

## Long Context

- DeepSeek-V3.1-Terminus: 43.5 (#88)
- Magistral Medium: 39.3 (#178)

| Benchmark | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| LMArena Longer Query | 1426 | 1295 |

## Writing & Preference

- DeepSeek-V3.1-Terminus: 61.3 (#87)
- Magistral Medium: 46.5 (#214)

| Benchmark | DeepSeek-V3.1-Terminus | Magistral Medium |
|---|---|---|
| LMArena Text | 1420 | 1255 |
| LMArena Creative Writing | 1404 | 1245 |
| LMArena Multi-Turn | 1411 | 1275 |

## FAQ

### Is DeepSeek-V3.1-Terminus better than Magistral Medium?

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 34.8 on the Noometry Index.

### Which is cheaper, DeepSeek-V3.1-Terminus or Magistral Medium?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; Magistral Medium lists at $2 and $5.

### Is DeepSeek-V3.1-Terminus or Magistral Medium better for coding?

DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 38.6 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek-V3.1-Terminus does, with 164K tokens against 128K.

### How many benchmarks do DeepSeek-V3.1-Terminus and Magistral Medium share?

13 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Magistral Medium has 22.